Two-stage phase II trials with early stopping for effectiveness and safety as well as ineffectiveness or harm

ABSTRACT

This article proposes new optimal and minimax designs, which allow early stopping not only for ineffectiveness or toxicity but also for sufficient effectiveness and safety. These designs may facilitate effective drug development by detecting sufficient effectiveness and safety at an early stage or by detecting ineffectiveness or excessive toxicity at an early stage. The proposed design has advantage over other designs in the sense that it can control the type I error rate and is robust against the real association parameter. Comparing to Jin's design, it is always advantageous in terms of expected sample size.

KEYWORDS:

• Early stopping
• minimax design
• optimal design
• phase II trials
• toxicity.

MATHEMATICS SUBJECT CLASSIFICATION:

•  

Acknowledgments

We thank the referees for the constructive comments which much improved the article. The Study is Supported by the Natural Science Foundation of Guangdong Province, China (2017A030313018).